What Is the Current Management of Osteoradionecrosis of the Temporal Bone Following Head and Neck Cancer Radiotherapy?

Osteoradionecrosis (ORN) of the temporal bone is a rare but serious late-stage complication of head and neck radiation. The most common symptoms include otorrhea, hearing loss, otalgia, tinnitus, and facial nerve palsy following radiation therapy for cancers of the nasopharynx, parotid, ear, skin, and central nervous system [1]. The pathophysiology of temporal bone ORN is related to the lateral/superficial location of the bone with respect to skin and soft tissue, which provide limited protection, reduced blood supply, and easy pathogen access via the Eustachian tube. While the true incidence of ORN of the temporal bone is unknown, a single-institution retrospective chart review of 82 patients undergoing surgery and subsequent radiotherapy for head and neck malignancies over a 10-year period from 2011 suggested an 8.5% incidence [2]. Though ORN of the temporal bone is rare, progression of the disease can contribute to intracranial complications (i.e., abscess, CNS infections) as well as cranial neuropathies, with even benign symptoms having a significant negative impact on patients' quality of life. There is wide variability regarding the management of this condition, with limited multi-institutional research to provide comprehensive evidence-based management. This, in part, is due to the rarity of the condition and the low sample size of patients included in current studies. As there are no prospective interventional trials for this disease, we are limited to treatments based upon case series. Case studies and retrospective chart reviews categorize intervention in ORN of the temporal bone by dividing the disease into local and diffuse ORN. Local disease is limited to the external auditory canal, with diffuse disease characterized by deeper invasion. Medical intervention includes topical antibiotics, in-office debridement, pain management, and new medications with adjunctive therapy (such as hyperbaric oxygen therapy), whereas surgical intervention can range from canalplasty, mastoidectomy, and lateral temporal bone resection with potential flap reconstruction. The goal of surgical management is to remove necrotic bone and protect all exposed bone. Of note, numerous studies acknowledge that there is no universal definition of successful treatment for temporal bone ORN; it is commonly defined as complete resolution of the symptoms of the disease following treatment [1, 3]. Conservative management, defined as topical antibiotics, in-office debridement, gentian violet application, and adequate analgesia, is first-line management for all types of temporal bone ORN [3]. Pentoxifylline-tocopherol (PENTO) was investigated as an adjunct to established conservative medical treatments, and research supports that PENTO may be an effective addition to standard conservative management in patients with localized ORN of temporal bone [4]. The treatment works by decreasing inflammation and fibroblast proliferation (inhibitory TNF-alpha effect) and reducing oxidative stress by stabilizing free radicals (vitamin E effect). It also increases the elasticity of red blood cells, improving vascularization where tissue perfusion is compromised. The addition of clodronate (a bisphosphonate) to PENTO or PENTOCLO has also been described as an effective treatment of mandibular ORN, but it is not as readily available in the United States, and due to concern for bisphosphonate-induced ORN, PENTO alone has been regularly studied and is effective against ORN. Hyperbaric oxygen (HBO) therapy is also another treatment option for temporal bone ORN. The therapy enhances oxygen delivery to hypoxic tissues, promoting angiogenesis, fibroblast proliferation, and collagen synthesis. There is variable evidence that this therapy is effective. Two randomized control trials and case series suggest that HBO can be an effective adjunct to conservative management in cases of localized disease, although these findings are not statistically significant as they are underpowered and did not routinely include a control group [5]. There is limited evidence on its effectiveness in diffuse disease, and HBO does not guarantee reduced wound breakdown when surgical intervention is needed. One randomized trial cited within the guidelines also showed that HBO was counterproductive in cases of mandibular ORN, possibly due to increased inflammation and oxidative stress [5]. Thus, current guidelines on the management of ORN for head and neck cancer patients treated with radiation therapy suggest that HBO can be a useful adjunct to PENTO and other medical management, though with weak evidence to support it [5]. Surgical management is primarily recommended in patients with diffuse disease, intractable pain, uncontrolled infections, cranial neuropathies, cholesteatomas, and when surgical intervention is safe for the patient. The overall goal for surgical management is to remove necrotic bone and obliterate any dead space to avoid further extension of the necrotic process. Mastoidectomy with vascularized flap closure has been shown to be effective in patients with extensive disease unresolved by medical management. When mastoid air cell destruction is present, lateral temporal bone resection with a vascularized flap reconstruction can be effective. Meta-analysis shows that surgical outcomes are better in patients younger than age 60 [1]. Temporal bone ORN is a rare but devastating complication of treatment for certain head and neck malignancies. Evidence suggests that medical management with topical antimicrobials, in-office debridement, PENTO, HBO therapy, and pain management may be an effective treatment for all patients, particularly those with localized disease. Surgical management should be reserved for those with diffuse disease, intractable pain, uncontrolled infections, cholesteatomas, and ages greater than 60 years. Further research needs to be done with randomized control trials and on novel methods to determine the best course of action for patients. This review includes a level 1, level 3a, level 3b, and two level 4 articles. The authors declare no conflicts of interest.